A collection of Claude Code skills, generalized for reuse across projects.
| Skill | What it does |
|---|---|
| cloudflare-worker-security | Harden a Cloudflare Workers edge surface: boundary auth, per-IP rate limiting, least-privilege bindings, allow-listed egress, scrubbed logs, no secrets in code. |
| git-history-discipline | Keep git history enterprise-grade and audit-ready: Conventional Commits, atomic changes, green-before-push, no secrets, no rewriting published history. |
| secure-agentic-build | Keep an AI-agent build pipeline trustworthy: defend against prompt-injection via untrusted content, malicious dependency suggestions, and secret exfiltration. |
Each skill lives in its own folder as a SKILL.md with YAML frontmatter
(name, description, tags) followed by a When to Use / Workflow /
Verification structure, matching the Claude Code skills format.
| Skill | What it does |
|---|---|
| neon | Overview of the Neon platform (Postgres, Auth, Data API, Object Storage, Functions, AI Gateway) and how to navigate its agent skills. |
| neon-postgres | Setup, connections, branching, autoscaling, pooling, and the Neon CLI/MCP/API for Neon Serverless Postgres. |
These two are unmodified copies of Neon's own official skills from neondatabase/agent-skills, included here for convenience — not originally authored in this repo.